Math.fround()

The Math.fround() static method returns the nearest 32-bit single precision float representation of a number.

JavaScript uses 64-bit double floating-point numbers internally, which offer a very high precision. However, sometimes you may be working with 32-bit floating-point numbers, for example if you are reading values from a Float32Array. This can create confusion: checking a 64-bit float and a 32-bit float for equality may fail even though the numbers are seemingly identical.

To solve this, Math.fround() can be used to cast the 64-bit float to a 32-bit float. Internally, JavaScript continues to treat the number as a 64-bit float, it just performs a “round to even” on the 23rd bit of the mantissa, and sets all following mantissa bits to 0. If the number is outside the range of a 32-bit float, Infinity or -Infinity is returned.

Syntax example: Math.fround(doubleFloat)